From owner-freebsd-current Mon Apr 3 07:24:41 1995 Return-Path: current-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id HAA07980 for current-outgoing; Mon, 3 Apr 1995 07:24:41 -0700 Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id HAA07918 for ; Mon, 3 Apr 1995 07:24:25 -0700 Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id AAA26234; Tue, 4 Apr 1995 00:18:47 +1000 Date: Tue, 4 Apr 1995 00:18:47 +1000 From: Bruce Evans Message-Id: <199504031418.AAA26234@godzilla.zeta.org.au> To: rgrimes@gndrsh.aac.dev.com, taob@gate.sinica.edu.tw Subject: Re: New installation notes Cc: freebsd-current@FreeBSD.org Sender: current-owner@FreeBSD.org Precedence: bulk >> But I were to do >> something like two 'dd if=blah of=/dev/null bs=65536' on files from >> each drive simultaneously, I should still be able to hit close to the >> maximum throughput, no? Probably not. On my system (DX2/66 VLB BT445C), one 7200rpm drive seems to hit a limit at 5.4MB/sec and reading another 3600rpm drive at the same time reduces the total throughput. >One needs to be very carefull when doing benchmarks to understand >what it is you are really measureing. And to attempt to make the >test measure what you really want it to measure. I image a carefully >written C program could do the ``repeat 1000 dd'' above and obtain >close to SCSI bus bandwidth if that is what you wanted to measure. >(I also suspect in this case the real bottleneck would be the disk >controller itself). Yes :-(. For the BT445C, repeatedly reading the same 512-byte block is almost exactly as slow as reading sequential 512-byte blocks. The speed depends on the drive too. I get about 180K/sec for a slow drive and 300K/sec for a fast drive. An Ultrastor 34F controller is 40% slower. Bruce